Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Harmful Query on JailbreakB
Loading...
0.67
ASR
NSPO
-1.1832
11.3259
23.835
36.3441
Dec 12, 2025
ASR
Updated 4d ago
Evaluation Results
Method
Method
Links
ASR
NSPO
Backbone=Qwen2.5-7B-In...
2025.12
0.67
NSPO
Backbone=Llama3-8B-Ins...
2025.12
1.33
BFPO
Backbone=Qwen2.5-7B-In...
2025.12
1.33
Llama3-8B-Instruct
Backbone=Llama3-8B-Ins...
2025.12
2
MoCAN
Backbone=Qwen2.5-7B-In...
2025.12
2
PeCAN
Backbone=Llama3-8B-Ins...
2025.12
2.33
DPO-H
Backbone=Qwen2.5-7B-In...
2025.12
2.33
DPO-S
Backbone=Llama3-8B-Ins...
2025.12
2.67
MoCAN
Backbone=Llama3-8B-Ins...
2025.12
2.67
BFPO
Backbone=Llama3-8B-Ins...
2025.12
2.67
DPO-S
Backbone=Qwen2.5-7B-In...
2025.12
2.67
Qwen2.5-7B-Instruct
Backbone=Qwen2.5-7B-In...
2025.12
3
W-DOOR
Backbone=Qwen2.5-7B-In...
2025.12
3.33
W-DOOR
Backbone=Llama3-8B-Ins...
2025.12
4.33
DPO-Mix
Backbone=Qwen2.5-7B-In...
2025.12
4.33
PeCAN
Backbone=Qwen2.5-7B-In...
2025.12
5.67
DPO-Mix
Backbone=Llama3-8B-Ins...
2025.12
6.67
SafeRLHF
Backbone=Llama3-8B-Ins...
2025.12
16.33
SafeRLHF
Backbone=Qwen2.5-7B-In...
2025.12
25
DPO-H
Backbone=Llama3-8B-Ins...
2025.12
47
Feedback
Search any
task
Search any
task